thinkhard.SI

Prompt repository Image & video

Show your design on a T-shirt, mug or package mockup

You have finished artwork and want a realistic photo of it on a product, for a store listing or a client presentation, with the design itself unchanged.

Registry no.
MED-015
Collection
Image & video
Laid out for
Image model, plain language
Score
10 out of 10
Blanks to fill
6
Added

Image & videoImage, plain

The prompt

Score 10 of 10, Excellent

Use it when You have finished artwork and want a realistic photo of it on a product, for a store listing or a client presentation, with the design itself unchanged.

  • {{use_case}}
  • {{product}}
  • {{placement}}
  • {{scene}}
  • {{light_direction}}
  • {{aspect_ratio}}
Full prompt and why it works
Prompt
Using the attached image of my design, create a realistic product photo for {{use_case}}, showing the design on {{product}}, {{placement}}. If I also attach a photo of the blank product, use it as the reference for the product's exact shape and color.

Keep the design exactly as it is in the attached file: the same shapes, colors, proportions, spacing and any text, letter for letter. Do not redraw, simplify, recolor or add to it. It should look printed on the product and follow its surface: curving around a mug, bending with the folds of fabric, or flat on the face of a box, with the material's texture showing slightly through, as real printing does.

The scene is {{scene}}. The product is the clear focal point, in sharp focus, with the background softer. The light is soft and natural, coming from {{light_direction}}, with gentle shadows that fall the same way across the product and the design, so the result reads as one photograph rather than a design pasted on top.

The image should look like a professional e-commerce photograph, photorealistic, in the style of a clean catalog shot, with true-to-life color, at an aspect ratio of {{aspect_ratio}}.

Leave out any other logos, labels, text, people, hands and watermarks. If part of the design would fall over a seam, handle or fold, move or scale the whole design to avoid it rather than distorting or cropping it. If any detail of the design is unclear at this size, such as fine lines or small text, keep it as it is in the file rather than inventing detail.

Variables

{{use_case}}
Where the image will be used, for example "the main photo of a print-on-demand listing" or "a client presentation".
{{product}}
The product, concretely, for example "a white ceramic 11 oz mug" or "a heather gray crew-neck T-shirt".
{{placement}}
Where the design goes and how big, for example "centered on the chest, about 10 inches wide".
{{scene}}
The setting, for example "on a wooden desk by a window, with a notebook beside it".
{{light_direction}}
Where the light comes from, for example "a window on the left".
{{aspect_ratio}}
The shape you need, for example 1:1 for most listings or 4:5 for social posts.

Why it works

  • What to avoid. The design must stay exactly as in your file, letter for letter, and nothing else branded may appear, so the mockup shows the product you will actually sell.
  • Background facts. Printing that follows the curve or the fabric, with shadows falling the same way across design and product, is what makes the result read as one photograph.
  • If unsure. A design that would land on a seam is moved or scaled rather than distorted, and fine detail is kept rather than invented. Check text and colors against your file before publishing.

How to use it

  1. Press Fill in variables and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/design-on-product-mockup

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "design-on-product-mockup"}, or just ask for it by name.